Every sales leader knows the feeling. You walk into a pipeline review with a number you believe in, and twenty minutes later, you're defending every line item to a CEO who just wants to know what's actually going to close.

HubSpot Sales Hub ends that conversation. Every deal, every rep's activity, and every buyer signal are all in one place and updated automatically. So your forecast is built on what's actually happening. And when you present that number, you can stand behind it.

Get a demo

👋 Hi, it’s Rohit Malhotra and welcome to the FREE edition of Partner Growth Newsletter, my weekly newsletter doing deep dives into the fastest-growing startups and S1 briefs. Subscribe to join readers who get Partner Growth delivered to their inbox every Wednesday morning.

If you’re new, not yet a subscriber, or just plain missed it, here are some of our recent editions.

_{Interested in sponsoring these emails? See our partnership options here.}

Subscribe to the Life Self Mastery podcast, which guides you on getting funding and allowing your business to grow rocketship. 

Previous guests include Guy Kawasaki, Brad Feld, James Clear, Nick Huber, Shu Nyatta and 350+ incredible guests.

Stripe's acquisition of OpenRouter is not a payments giant chasing AI hype at the top of the market. It is Patrick Collison paying up for the one position that still compounds in AI - the switchboard every application already routes through - before the rest of the market prices it.

On the surface: Stripe pays more than $7 billion for a startup founded in 2023 that closed a $113 million Series B just three months ago at $1.3 billion. That is over 5x the number the venture market set that spring. CapitalG led the round. Ninety days later, $1.3 billion looks small.

But OpenRouter did not balloon on froth. It became plumbing. Eight million users. Four hundred–plus models. The neutral layer between every AI app and every provider — and Stripe already ran its payments.

That is not overpaying. It is buying the tollbooth before everyone admits they need it.

In 2023, Alex Atallah, fresh off co-founding OpenSea, built a single gateway that let developers reach hundreds of AI models through one integration instead of wiring up each provider separately. That became OpenRouter. Three years later, Stripe is paying more than $7 billion for it. The reason isn't the technology. It is the position: the neutral switchboard sitting between every AI application and the models it runs on.

The asset compounded fast. OpenRouter launched as a way to pool developers together for cheaper API credits, then became the routing layer the AI economy quietly depends on. Today it reaches 8 million users and provides access to more than 400 models from OpenAI, Anthropic, and the rest. In May 2026 it raised a $113 million Series B at a $1.3 billion valuation, led by CapitalG.

That is the asset Stripe wants. It already handled OpenRouter's payments, so it watched the transaction flow up close. Stripe sits between businesses and money; OpenRouter sits between applications and models. Same shape, one layer up. As agents and AI apps generate their own transactions, model access, billing, and usage-based pricing collapse into one flow. Ninety days after that $1.3 billion round, Stripe is paying more than five times it.

Stripe is paying more than $7 billion, and the reported terms carry none of the machinery you'd expect on a number that size. Three months earlier, the entire company was valued at $1.3 billion. So the price is not built off ARR, and not off a balance sheet stripped down to cash. It is built off a position. Five times the last mark, set in ninety days.

The number that matters is what is missing. There is no distress. No down round. No fallen unicorn the venture market can no longer clear. OpenRouter raised at $1.3 billion in May 2026 with CapitalG leading, and every signal pointed up. Stripe is not buying a discount. It is paying a premium, in cash, to take the asset off the table before the next round priced it higher.

And what it is buying is not revenue to harvest. It is a chokepoint to own. The neutral layer every AI application already routes through, 8 million users deep, 400 models wide. Stripe already cleared the payments underneath it. The $7 billion buys the switchboard itself, and the token flow that has barely started.

Three numbers do the explaining here.

One. Stripe is paying more than 5x the last valuation, set ninety days earlier. In May 2026 OpenRouter raised $113 million at $1.3 billion. By August the price is north of $7 billion. Nothing in the business changed that fast. What changed is what the position is worth once a buyer decides it is load-bearing. That gap is not an OpenRouter story. It is the AI-infrastructure repricing made literal, the moment the switchboard gets valued like the switchboard.

Two. The multiple has no denominator you can point to. There is no clean revenue figure, no ARR anchor, no earnout tied to model volume. OpenRouter takes a thin cut of the tokens routed through it, so the value isn't this year's margin. It is the flow itself, 8 million users and 400 models wide, and the assumption that it compounds as tokens become the medium every AI company trades in.

Three. Stripe already ran the payments underneath OpenRouter. It watched the transaction volume up close, priced the growth off its own data, and paid a premium anyway. This isn't a buyer reaching for a number. It is the one party positioned to know exactly what the number should be, deciding it was worth 5x.

Stripe is paying the full number up front for what OpenRouter is today. Most acquirers at this size hedge, tying half the price to growth that may not arrive. Stripe didn't. It watched the token flow through its own payment rails, decided the position was real, and paid for it in one stroke. That's not caution. It's conviction.

And this isn't a brand to run for margin. It's a layer to absorb. Where a holding company keeps the logo and strips the body, Stripe does the opposite, wiring OpenRouter into the flow it already owns. Payments underneath, model routing on top, the same request paying for compute and clearing the transaction in one motion. OpenRouter, A Stripe Company, is the phrasing, and the direction of travel is integration, not harvest.

Funded by the most valuable private company in fintech, with the payment relationship already in place. The precedent isn't a wind-down. It's Stripe buying the rails before the traffic fully arrives.

Every AI app hits the same wall.

Not the model. The meter.

You wire up OpenRouter, route a million requests, and then you have to bill for compute you don't control, price it per token, and settle it against real dollars. That plumbing is brutal. It's the exact problem Stripe spent over a decade solving for the last internet.

Here's what most people miss.

Stripe already ran OpenRouter's payments.

It didn't learn about the token flow from a pitch deck. It watched it clear, request by request, on its own rails.

So this isn't a payments company reaching into AI. It's the one operator that already sees both sides of the transaction: the dollars going in, the tokens going out.

8 million developers. 400 models. One switchboard.

And underneath it, the company that already knows how to move money at that scale without dropping a cent.

The operator logic is simple. Whoever meters the tokens and settles the dollars in a single motion owns the layer.

Stripe was already halfway there.

It just bought the other half.

Stripe is wagering on four things being true at once.

One. Tokens become a permanent medium of exchange, not a phase. The thesis even has a slogan now: tokens are the new dollars. Stripe is betting the flow of intelligence between companies keeps compounding, which is why it paid for the switchboard and not for a single year of revenue.

Two. Multi-model beats lock-in, and stays that way. 8 million developers route across 400 models precisely because no lab stays ahead forever. Stripe is betting no single model ever wins hard enough to make the router unnecessary.

Three. Owning the layer beats taxing it. Stripe already processed OpenRouter's payments and could have kept clipping the transaction from the outside. Instead it bought the routing itself. It is betting the value is in owning the chokepoint, not renting a slice of it.

Four. The position compounds faster than the price. There is barely any revenue to underwrite here. Stripe is betting that agents and AI apps start generating their own transactions, that model access, billing, and usage pricing collapse into one flow, and that owning that flow at $7 billion looks cheap once everyone realizes they need it.

Every acquisition has a scoreboard. Here's who reads red on this one.

The rival gateways lose first. Every other AI router, Portkey, LiteLLM, Cloudflare's gateway, now competes against a switchboard owned by one of the most powerful companies in fintech, with the payment rails already bolted on. The independent middle just got a lot lonelier.

The model labs lose next. OpenRouter's entire job is to make models interchangeable, ranked by price and latency. Every request that clears through it strips a lab of its direct customer and its pricing power. OpenAI and Anthropic become a dropdown menu, and the dropdown just got a $7 billion owner.

The billing startups lose too. Metronome, Orb, the usage-based metering crowd, all watch Stripe annex the token meter itself, the exact layer they were counting on to be theirs.

But the real loser isn't at this table. It's neutrality. OpenRouter's whole value was being the Switzerland of models, unowned and indifferent to who wins. Now the switchboard answers to one company with its own stack to sell.

The winner is the buyer that owns both the dollars and the tokens. In an intermediated economy, the toll collector always wins.

Three threads worth pulling.

One. Stripe just declared the AI economy runs on rails, and it intends to own them. It processes the dollars; now it owns the layer that routes the tokens. The signal isn't OpenRouter. It's that the winning bet in AI may not be a model at all, but the infrastructure sitting between the models and the money. Watch which pipes get bought next.

Two. The venture market and the acquisition market have decoupled, in the other direction. Airtable exited below its last private mark. OpenRouter exits at more than five times its mark from ninety days ago. Growth-era SaaS gets discounted; AI position gets a premium the venture market hadn't caught up to. The signal is that private marks and clearing prices are unhinged in both directions now. Watch how fast the next AI infra round reprices.

Three. Revenue stopped being the thing you buy. Stripe paid $7 billion for a company with almost no revenue to underwrite, because the token flow it sits on is the asset. The multiple has no denominator. The signal is that in AI, position is being priced above profit, the way audience once was. Watch how many 2026 deals buy the flow and ignore the P&L.

The bull case. Stripe bought the neutral switchboard the entire AI economy already routes through, 8 million developers and 400 models deep, with the payment rails underneath it already its own. If tokens become the new dollars, it just bought the clearinghouse for both, and $7 billion looks like a down payment on the next layer of the internet.

The bear case. It paid more than five times a ninety-day-old mark for a company with almost no revenue and a wide open front door. Routers are thin, replaceable, and margin-poor. If a lab cuts off access, or the model race consolidates to two winners, the switchboard has less to switch between and less pricing power to defend.

What to watch: the take rate. A router that clips a fraction of every token can compound into infrastructure, or get competed to zero. If Stripe can hold its cut as the flow scales, the position is real ie. and permanent.

Here is my interview with Mike Zisman, the founder and Executive Chairman of Golf Genius Software, the leading tournament management platform serving 11,000+ golf clubs across 62 countries

If you enjoyed our analysis, we’d very much appreciate you sharing with a friend.

Here are the options I have for us to work together. If any of them are interesting to you - hit me up!

And that’s it from me. See you next week.

What do you think about my bi-weekly Newsletter? Love it | Okay-ish | Stop it